Although a considerable body of research has focused on the neural substrates of social separation, relatively little is known about the neurobiology of social attachment. This proposal focuses on a monogamous rodent, the prairie vole, which forms selective, enduring social attachments or pair bonds. Several lines of evidence implicate central oxytocin (OT) pathways in prairie vole formation. OT increases and a selective OT antagonist decreases mating-induced pair bonding in female prairie voles as measured by the development of a partner preference. Blockade of OT receptors specifically in the nucleus accumbens inhibits development of a partner preference in female prairie voles. OT receptors are not found in this region in related vole species that do not form pair bonds. In prairie voles, OT receptors in the nucleus accumbens show a surprising individual variability. In approximately 20% of prairie voles, OT receptor mRNA or binding is virtually absent in this region, while OT receptors in other brain regions remain unaffected. We propose four new studies to investigate the importance of OT receptors in the nucleus accumbens for the development social attachments. The first study increases regional OT receptor expression in both prairie voles and non-monagamous montane voles, which do not form pair bonds. The second study explores the individual variation in nucleus accumbens OT receptors in prairie voles by comparing females with high and low levels of binding for regional gene induction and partner preference formation in response to OT. A third study investigates potential development mechanisms for the individual variation, based on a recent discovery that rats raised under different rearing conditions have different levels of OT receptor binding in adulthood. Finally, a fourth study explores molecular mechanisms for the individual differences in OT receptor expression. Taken together, these studies not only define a key step in the neurobiology of pair bonding, they may provide a molecular, cellular, and behavioral understanding of individual variation in a natural animal model system. The ultimate goal of this research is to provide new approaches to clinical disorders characterized by deficits in attachment, including autism and schizophrenia.